Bidirectional programming languages have been proposed as a practical approach to the view update problem. Programs in these languages, often called lenses, can be read in two ways— from left to right as functions mapping sources to views, and from right to left as functions mapping updated views back to updated sources. Lenses address the view update problem by making it possible to define a view and its associated update policy together. One issue that has not received sufficient attention in the design of bidirectional languages is alignment. In general, to correctly propagate an update to a view, a lens needs to match up the pieces of the edited view with corresponding pieces of the underlying source. Unfortunately, existing bidirectional languages are extremely limited in their treatment of alignment—they only support simple strategies that do not suffice for many examples of practical interest. In this paper, we propose a novel framework of matching lenses that extends basic lenses with new mechanisms for calculating and using alignments. We enrich the types of lenses with “chunks” that identify the locations of data that should be re-aligned after updates, and we formulate refined behavioral laws that capture essential constraints on the handling of chunks. To demonstrate the utility of our approach, we develop a core language of matching lenses for string data, and we extend it with primitives for describing a number of useful alignment heuristics.